Thursday, July 9, 2020

Creating an ACFS filesystem on Exadata Cloud Service

Below is an example on how to create an ACFS filesystem on an Exadata Cloud Service quarter rack.  This example creates a 2TB ACFS in the RECO1 disk group and is mounted at /scratch.

[opc@exacsphx-xyzn1 ~]$ sudo -s
[root@exacsphx-xyzn1 opc]# su - grid
Last login: Thu Oct 17 09:38:15 EDT 2019
[grid@exacsphx-xyzn1 ~]$ asmcmd
ASMCMD> volcreate -G RECOC1 -s 2000G upload_vol
ASMCMD> volinfo -G RECOC1 upload_vol
Diskgroup Name: RECOC1

         Volume Name: UPLOAD_VOL
         Volume Device: /dev/asm/upload_vol-58
         State: ENABLED
         Size (MB): 2048000
         Resize Unit (MB): 64
         Redundancy: HIGH
         Stripe Columns: 8
         Stripe Width (K): 1024

ASMCMD> exit
[grid@exacsphx-xyzn1 ~]$ logout

[root@exacsphx-xyzn1 opc]# /sbin/mkfs -t acfs /dev/asm/upload_vol-58
mkfs.acfs: version                   =
mkfs.acfs: on-disk version           = 46.0
mkfs.acfs: volume                    = /dev/asm/upload_vol-58
mkfs.acfs: volume size               = 2147483648000  (   1.95 TB )
mkfs.acfs: Format complete.
[root@exacsphx-xyzn1 opc]# /u01/app/ add filesystem -d /dev/asm/upload_vol-58 -g RECOC1 -v upload_vol -m /scratch
[root@exacsphx-xyzn1 opc]# /u01/app/ start filesystem -d /dev/asm/upload_vol-58

[root@exacsphx-xyzn1 opc]# df -h
Filesystem                    Size  Used Avail Use% Mounted on
devtmpfs                      355G     0  355G   0% /dev
tmpfs                         709G  1.5G  707G   1% /dev/shm
tmpfs                         355G  3.0M  355G   1% /run
tmpfs                         355G     0  355G   0% /sys/fs/cgroup
/dev/mapper/VGExaDb-LVDbSys1   24G   11G   12G  47% /
/dev/xvda1                    488M   82M  381M  18% /boot
/dev/xvdi                     1.1T   69G  959G   7% /u02
/dev/mapper/VGExaDb-LVDbOra1   20G  7.4G   12G  40% /u01
/dev/xvdb                      50G  9.3G   38G  20% /u01/app/
/dev/asm/acfsvol01-58         800G   39G  762G   5% /acfs01
tmpfs                          71G     0   71G   0% /run/user/2000
tmpfs                          71G     0   71G   0% /run/user/0
/dev/asm/upload_vol-58        2.0T  4.6G  2.0T   1% /scratch
[root@exacsphx-xyzn1 opc]#

Tuesday, July 7, 2020

Autonomous Database with HammerDB

Summary of the basic steps to configure HammerDB with Oracle's Autonomous Database either shared or dedicated.
  1. Download and install Oracle Database Client on client machine which is hosting HammerDB
  2. Optionally it is recommended practice to patch the client to the latest version, today that is 19.7
  3. set environment variable Oracle_HOME
  4. Download the wallet for the database you will be using with HammerDB
  5. Unzip the wallet , modify sqlnet.ora and set TNS_ADMIN to your wallet
  6. Modify oracle.xml file in your hammerdb/config directory
  7. Verify above settings
My Environment:
  • Windows jump server
  • Linux Compute VM to host hammerdb
  • ADB on Dedicated Quarter Rack
  • ADB on Shared Infrastructure
  • all on same VCN

#1   Download and install latest Oracle Database Client which currently is 19.3 onto your linux compute VM.

#2 Skipping this step

#3   Set the environment variable ORACLE_HOME to your client on .
       i.e.   export ORACLE_HOME=/home/opc/app/opc/product/19.0.0/client_1

#4   Download the wallet  for the database you will be using onto  your linux compute VM


#5  Unzip the wallet onto your linux compute VM and set TNS_ADMIN to this directory
        export TNS_ADMIN=/home/opc/wallet_phxhammer1


Also modify the sqlnet.ora to point to this directory.

#6  Either modify the oracle.xml in  the config directory within the hammer directory or use the GUI.   Make the following changes in config.xml:
  • system_user = admin
  • system_password = your_admin_password
  • instance = use one of the defined services   i.e.  phxhammer1_tp
  • tpcc_user = tpcc
  • tpcc_pass = insert password for the tpcc user
  • tpcc_def_tab = DATA
  • tpcc_def_temp= TEMP 
Here is a portion of the config file:

#7 start HammerDB  and verify connection